US 7995564 B1
In a bearer-independent core network, the switching and transport of the bearer streams is separated from the processing of call sessions that control the bearer streams. The switching and transport of the bearer streams is executed by media gateways and packet switches in the switching and transport layer of the network. The processing of calls sessions that control the bearer streams is executed by call servers in the call control layer of the network. The switching and transport layer of the network may be implemented using ATM or IP technology. In order to redirect the signaling links to a backup call server, a signaling gateway is used. Each media gateway includes a signaling gateway function. The signaling gateways redirect the signaling links by encapsulating each signaling message in a new packet in one embodiment.
1. A method for transferring call control to a backup call server, comprising:
monitoring a primary call server to determine an active or inactive state of the primary call server; and
upon receipt of an inactive state for the primary call server, forwarding signaling messages from a signaling gateway of a plurality of signaling gateways to a backup call server by encapsulating the signaling messages in a data packet with a destination address of the backup call server, wherein each of the plurality of signaling gateways has an assigned backup call server.
2. The method of
3. The method of
4. The method of
5. The method of
6. The method of
7. The method of
8. A signaling gateway for a cellular network communicatively coupled to a destination switching element and to at least one home location register (HLR), comprising:
a memory for storing computer instructions that define the operational logic of the signaling gateway, wherein the computer instructions include logic for:
receiving call signaling messages from the at least one HLR;
determining whether the destination switching element is in an inactive state; and
if the destination switching element is in an inactive state,
determining whether a first backup switching element corresponds with the destination switching element; and
if the first backup switching element corresponds with the destination switching element, transparently forwarding the call signaling messages to the first backup switching element by encapsulating the signaling message in a data packet with a destination address of a backup call server.
9. The signaling gateway of
determining whether a second backup switching element corresponds with the destination switching element, and
if the second backup switching element corresponds with the destination switching element, transparently forwarding a second group of call signaling messages to the second backup switching element.
The present U.S. Utility patent application claims priority pursuant to 35 U.S.C. §120 as a continuation to the following U.S. Utility patent application, which is hereby incorporated herein by reference in its entirety and made part of the present U.S. Utility patent application for all purposes:
1. U.S. Utility application Ser. No. 10/615,260, entitled “Geographic Redundancy for Call Servers in a Cellular System based on a Bearer Independent Core Network,”, filed Jul. 8, 2003, now issued as U.S. Pat. No. 7,532,568, on May 12, 2009, that claims priority pursuant to 35 U.S.C. §119(e) to the following U.S. Provisional Patent Application, which is hereby incorporated herein by reference in its entirety and made part of the present U.S. Utility Patent Application for all purposes:
a. U.S. Provisional Application Ser. No. 60/394,632, filed Jul. 9, 2002, expired.
1. Technical Field
The present invention relates to communication networks and, more particularly, to wireless communication networks.
2. Related Art
Initially, communication systems, and more particularly, wireless communication systems, would route calls through the actual trunks that carry the calls. Stated differently, the trunks would be used to establish the link from the calling party to the called party without knowing the availability of the called party. The impact of this approach was that a trunk would be tied up for call routing purposes even if the call was not eventually connected. While the telephone resources would be released as soon as it was clear that the called party was not available, the wasteful allocation of resources for calls that could not be connected, in aggregate, resulted in significant inefficiencies.
Accordingly, telecommunication networks evolved to include signaling systems to facilitate the routing of call setup signals without having to tie up actual communication trunks. More generally, the purpose of a signaling system is to transfer control information between elements in a telecommunications network. The elements may comprise switches, operation centers and databases. For example, in some telecommunication networks, the elements comprise base station transceiver sets (BTSs), base station controllers (BSCs), mobile switching centers (MSCs), home location registers (HLRs), etc. Thus, signaling systems were originally developed to establish connections between telephone offices and the customer premises equipment (namely, the telephones) in order to transport voice traffic through a voice-oriented telephone network. To explain in simple terms, the signaling systems use a separate channel for conducting signaling control messages.
In operation, a mobile station, such as mobile station 38, generates call setup signals through a BTS to a base station controller (BSC), collectively BS 14, in this example, which then generates calling signals to MSC 18. For the present example, assume that mobile station 38 is placing a call to mobile station 42. Neither BS 14 nor MSC 18 know what BS is serving mobile station 42 or what MSC is serving mobile station 42. Accordingly, MSC 18 produces call setup signals to a G-MSC 34 which queries a home location register (HLR) 30 to determine a location of mobile station 42. More specifically, HLR 30 returns an identifying MSC, which here is MSC 26, to G-MSC 34.
In some cellular networks, HLR 30 would merely communicate with the destination MSC as a part of the call setup procedures. Accordingly, in those embodiments, HLR 30 may communicate with MSC 26 to advise it that MSC 18 is attempting to establish a call to a mobile station served by MSC 26, as well as identifying information to enable MSCs 18 and 26 to communicate to setup the trunk for the voice traffic. In other networks, for example, in global systems for mobile communications (GSM) networks, the G-MSC, here G-MSC 34, communicates with HLR 30 to determine a destination MSC. Accordingly, in these embodiments, HLR 30 returns the identity of MSC 26 to G-MSC 34 which, in turn, forwards the response to MSC 18. The MSC 18 then communicates with MSC 26 to further set up the call. In the example shown, MSC 18 communicates with MSC 26 by way of G-MSC 34. Accordingly, the call connection may be established for a mobile station 38 through each of the communication elements to mobile station 42.
As the dual networks evolved for carrying user traffic and signaling traffic, the focus of many designs was to also provide redundancy, not only for the user traffic, but also for the signaling links. Thus, it is common to have a redundant signaling link, perhaps in a different cable bundle or, in a fiber optic ring, routed in an opposite direction, to provide an ability for control signaling to propagate through the network to establish a call despite a link or element failure. Heretofore, however, no geographically separated redundancy has been provided for the endpoints of the signaling network that carries the signaling traffic. More particularly, in phone networks, and especially in cellular phone networks, no geographically separated redundancy has been provided for the mobile switching centers and the gateway mobile switching centers.
Typically, the MSCs are consolidated at a switching center geographically distant from the cellular networks they control. One problem associated with consolidating the switching function is a geographically localized event, such as a tornado or other natural/man-made disaster will disrupt communications over a large area for any network element for which geographically separated redundancy is not provided.
The control or signaling network includes switching and transport of signaling streams and is separated from the processing of the call sessions that control the bearer streams. The switching and transport of the bearer streams is executed by media gateways and packet switches in the switching and transport layer of the network. The processing of call sessions that control the bearer streams is executed by call servers in the call control layer of the network. The switching and transport layer of the network typically uses asynchronous transfer mode (ATM) or Internet protocol (IP) technology.
One advantage of having bearer-independent core networks is that call servers can handle calls for media gateways that are distributed across a very large geographic area. One shortcoming of this approach, however, is that the risk resulting from the failure of a call server due to a power outage or natural disaster at the call server site, as mentioned above, affects communication services, and more particularly, wireless communication services, over a larger geographic area than the failure of a traditional cellular mobile switching center. Accordingly, what is needed is a call server that may be coupled and formed to act as a backup server in the event of a failure of a primary call server even if the backup call server is geographically dispersed from the primary call server.
A signaling gateway is used to redirect signaling links from a primary call server to a backup call server in the event of a failure of the primary call server. The signaling gateway redirects signaling messages intended for the primary call server by encapsulating each signaling message in a new packet containing an address of a backup call server in one embodiment of the invention. In another embodiment of the invention, a mapping table is updated to automatically direct data packets to the backup call server. In either embodiment, the packet has the appropriate destination address of the backup call server when the primary call server is failed (inactive). For example, if a primary MSC or gateway-MSC (G-MSC) has failed, then the signaling gateways that form a ring around a network of MSCs and G-MSCs redirect portions of a primary signaling stream to backup call servers. One primary benefit of this technique is that it minimizes SS7 network disruption resulting from links being bounced. Another benefit is that SS7 peer addressing does not need reconfiguring when a geographically redundant call server fails. Generally, MSCs and G-MSCs are similar devices that, for the purpose of the described functions in a particular example, perform slightly different tasks during call setup. For example, the G-MSC is an MSC that communicates with an HLR to obtain location information for a mobile.
In one embodiment of the invention, one backup call server performs all backup functions for a primary call server. In an alternate embodiment of the invention, however, portions of the signaling links are directed to different backup call servers. From an external standpoint, meaning external to the signaling gateways, the change from a primary call server to a backup call server is transparent.
As stated before, in a bearer-independent core network, the switching and transport of the bearer streams is separated from the processing of call sessions that control the bearer streams. The switching and transport of the bearer streams is executed by media gateways and packet switches in the switching and transport layer of the network. The processing of calls sessions that control the bearer streams is executed by call servers in the call control layer of the network. In order to redirect the signaling links to a backup call server, however, a signaling gateway is used. Each media gateway often includes a signaling gateway function though the media and signaling gateway functions may be separated. In the described embodiments wherein a signaling gateway is described, such signaling gateway may be associated with or separated from the media gateway. The described embodiments primarily related to the signaling gateway functionality though the inventive concepts may also be employed with media gateways for media transport functions. The signaling gateways redirect the signaling links by encapsulating each signaling message in a new packet. The new packet has the address of the primary call server during normal conditions or the address of the backup call server when the primary call server has failed. Messages are encapsulated in a packet format defined by an packet based (such as an IP-based) signaling protocol. This enables the signaling messages to be routed between the call servers and signaling gateways over an IP network. When geographic redundancy is introduced into a network in this manner, each media gateway homed on the primary call server is assigned a backup call server. Every call server may act as both a primary call server and a backup call server in the described embodiments of the invention. Additionally, a call server may act as a backup call server for media gateways homed on (coupled proximately to) more than one primary call server. Each media gateway homed on a particular call server may be backed up by a different backup call server in order to distribute the impact of a failover condition over several call servers.
In one embodiment of the present invention, a plurality of signaling gateways form a network ring around a plurality of call servers to create geographic redundancy among the call servers there within. Each media gateway associated with a primary call server, or geographically coupled proximate to the primary call server, is assigned at least one backup call server. Within the ring of media gateways, every call server may act as both a primary call server and a backup call server in the described embodiment. Moreover, a call server may act as a backup call server for media gateways that are homed on more than one primary call server. Additionally, each media gateway homed on a particular call server may be backed up by a different backup call server in order to distribute the impact of a failover condition over several call servers.
Further, certain provisioned data in each primary call server is replicated in its backup call server to facilitate fast and seamless backup processing. This data replication may be performed by the operations, administration and maintenance system for the network. It may also be performed through any other known means to one of average skill in the art. In one embodiment of the invention, the replicated data for the primary call server to the backup call server includes translation data for routing calls, integrated services digital network user part (ISUP) routes, SS7 point codes, location area codes, media gateway inventory with an indication of active/inactive status, and signaling gateway inventory with an indication of active/inactive status for each.
Until an actual failover occurs, meaning that a signaling gateway fails to receive a “heartbeat” from the primary call server, this data is tagged as “inactive” in the backup call server. While tagged as “inactive”, the backup call server does not use this data to process calls. Upon a specified error condition, for example, failure to receive a heartbeat, the associations within the media gateway of inactive are changed to active. Otherwise, the inactive associations from the backup call server's perspective will not cause alarms at the backup call server. When the fail over occurs, however, and the data is tagged as “active” in the backup call server, the backup call server immediately begins to utilize the data to process the calls. Accordingly, the backup call processing happens very quickly and is as seamless as possible.
In general, for the example of
In operation, each of the signaling gateways receives a heartbeat signal that is produced by each of the G-MSCs of the network. Accordingly, G-MSC 34, 72 and 76 each produce a heartbeat to the signaling network ring 52. Upon failure of any one of the G-MSCs 34, 72 or 76, each of the signaling gateways is operable to perform specified backup data packet routing for the signaling network. For example, we will presume that G-MSC 34 has failed to produce its heartbeat signal within a specified window and it is therefore considered “inactive”. As MSC 18 produces signaling messages, or data packets, containing an address of G-MSC 34 to signaling gateway 56, signaling gateway 56 will route the signaling messages, using backup signaling paths 98 and 94, to G-MSC 76 and G-MSC 72. Since G-MSC 76 and G-MSC 72 are operable as backup call servers to G-MSC 34, then signaling gateway 56 will, either by remapping the signaling messages to the designated backup call servers, or by adding new header address information to the signaling messages, route the signaling messages to the corresponding backup call server.
In the described embodiment of the invention, the primary call server, here G-MSC 34, is backed up by a plurality of G-MSCs or call servers, such as G-MSC 76 and G-MSC 72. It is understood that if the signaling network ring encompasses a greater number of call servers, then a greater number of call servers may operate as backup call servers for G-MSC 34. For the sake of simplicity, however, the present invention shows a signaling network ring that provides adaptive call routing for backup purposes for only three call servers, namely, G-MSC 34, G-MSC 72 and G-MSC 76. The designation as to what call server provides backup functionality is merely left to the system designer or system administrator of the network. As mentioned before, the routing of signaling messages to the backup call servers may be done either through a mapping scheme or by adding headers containing addresses of the backup call servers for the signaling messages. Both of these approaches are known by those of average skill in the art.
Continuing to examine communication network 50, in the situation where G-MSC 34 has failed to produce a heartbeat in a specified window of time, a plurality of external communication elements may originate signaling messages intended for G-MSC 34. For example, MSC 68, MSC 92, MSC 26 or MSC 96, may each have signaling messages intended for G-MSC 34. In one embodiment of the present invention, each backup call server processes a specified group of signaling messages for each of the MSCs. Each signaling gateway proximately coupled to a MSC is therefore provisioned to perform the specified backup call routing in the event of a failure of a specified G-MSC, such as G-MSC 34. In an alternate embodiment of the invention, each backup call server processes a specified type of signaling message. Accordingly, G-MSC 76 may provide backup call processing for first and second types of signaling messages, while G-MSC 72 provides backup call processing third, fourth and fifth types of signaling messages. As stated before, the specific implementations are function of system administration.
In operation, each signaling gateway is provisioned to perform specified backup routing to the corresponding backup call server. Thus, signaling gateway 56 will route, in the described example, signaling messages only to G-MSC 76 upon the failure of G-MSC 34. Signaling gateway 88, however, will route signaling messages to both G-MSC 34 and G-MSC 72 upon the failure of G-MSC 76. Signaling gateway 84, like signaling gateway 56, only routes signaling messages to one backup call server, namely, G-MSC 76. Accordingly, signaling gateway 84 will route signaling messages over backup signaling path 128 to G-MSC 76 upon the failure of G-MSC 72. As described before, each signaling gateway may be provisioned to route signaling messages to a corresponding backup call server by a system administrator in a desired manner. Moreover, the backup call routing may be done either through a mapping scheme or through a layered addressing scheme. The layered addressing scheme, in one embodiment, includes encapsulating a signaling message with a new header that contains the backup call server address.
The operation of the network of
When the primary call server transitions back to an active state, the heartbeat signal will be sensed by the plurality of signaling gateways. Once it is determined the primary call server has transitioned to the active state, the signaling messages are forwarded to the primary call server (step 240). In one embodiment, the primary call server processes all the signaling messages it would have processed prior to transitioning to the inactive state (step 244). In an alternate embodiment, the primary call server processes different signaling messages from what it would have processed prior to transitioning to the inactive state (step 248).
The invention disclosed herein is susceptible to various modifications and alternative forms. Specific embodiments therefore have been shown by way of example in the drawings and detailed description. It should be understood, however, that the drawings and detailed description thereto are not intended to limit the invention to the particular form disclosed, but on the contrary, the invention is to cover all modifications, equivalents and alternatives falling within the spirit and scope of the present invention as defined by the claims.